ModelCap
Current AI model comparison
Exact identities · public evidence
IBM Granite
granite 3.3 2b instruct
Position
#248
Index
7.8
Estimated
publisher-corpus-prior over 209 held-out anchors (8%); opens above measured predecessor ibm-granite/granite-3.1-2b-instruct on same-family ordering evidence (30%); launch card against 4 resolved peers on 5 rows (62%); exceeds every named peer on 2 of 5 rows; optimism haircut 1.61 from cross-lab-probe; shrunk 1 up toward the measured corpus
Comparison side 1
Microsoft
UserLM 8b
Position
#246
Index
8.0
Inherited
finetune of meta-llama/Llama-3.1-8B · finetune (100%)
Comparison side 2
Output / 1M
Unavailable
vs Unavailable
Context
—
vs —
Weights
Open weights
vs Open weights